Social Protection for the Poor and Poorest in Developing Countries: Reflections on a Quiet Revolution

Armando Barrientos and David Hulme1 March 2008 BWPI Working Paper 30

Creating and sharing knowledge to help end poverty

1 School of Environment and Development, University of Manchester armando.barrientos@manchester.ac.uk david.hulme@manchester.ac.uk Brooks World Poverty Institute ISBN : 978-1-906518-29-5

www.manchester.ac.uk/bwpi

2

Abstract

The concept and practice of social protection in developing countries has advanced at

an astonishing pace over the last decade or so. There is a growing consensus around

the view that social protection constitutes an effective response to poverty and

vulnerability in developing countries, and an essential component of economic and

social development strategies. This paper argues that the rise of social protection

constitutes a response to global factors, but with considerable regional diversity. The

paper examines the factors determining the future course of social protection and

identifies urgent research needs.

Keywords: social protection, vulnerability, human capital, productive assets, rights-

based approach.

David Hulme is Professor in Development Studies at the Institute for Development

Policy and Management at the University of Manchester and is Director of the Chronic

Poverty Research Centre. He is an Associate Director of the Brooks World Poverty

Institute.

Armando Barrientos is an Associate Director of the Brooks World Poverty Institute.

He is also a Senior Researcher with the Chronic Poverty Research Centre, leading

research on the Insecurity, Risk and Vulnerability research theme.

3

1. Introduction

The concept and practice of social protection in developing countries have advanced at

an astonishing pace over the last decade or so. There is a growing consensus around

the view that social protection constitutes an effective response to poverty and

vulnerability in developing countries, and an essential component of economic and

social development strategies.i Social protection is now better grounded in development

theory, and especially in an understanding of the factors preventing access to economic

opportunity and sustaining persistent poverty and vulnerability. The initially dominant

conceptualisation of social protection as social risk management is been extended by

approaches grounded in basic human needs and capabilities. Social protection practice

has also changed from a focus on short term social safety nets and social funds to a

much broader armoury of policies and programmes that combine interventions

protecting basic levels of consumption among poor and poorest households; facilitating

investment in human capital and other productive assets which provide escape routes

from persistent and intergenerational poverty; and strengthening the agency of those in

poverty so their capability to overcome their predicaments are increased (Barrientos and

Hulme, 2008).

While programme design issues have dominated discussion of social protection, issues

of scale are of much greater significance. The rapid introduction of social protection

programmes based on income transfers has resulted in a steep rise in coverage. New

forms of social assistanceii introduced in the last decade now reach in excess of 150

million poor households in developing countries, with perhaps half a billion people

benefiting. Notable recent initiatives include: South Africa’s Child Support Grant,

implemented in 2003 and now reaching 7.2 million children; the Minimum Living

Standards Scheme in China initiated in the late 1990s and reaching 22.4 million by

2006; Mexico’s Oportunidades started in 1997 and now reaching more than 5 million

households; Bolsa Familia in Brazil with coverage of 12 million households; Indonesia’s

Safety Net Scheme introduced in 2005 and planned to reach 15 million households; and

India’s National Rural Employment Guarantee Scheme expected to reach 26 million

households during 2008.iii This scaling up has changed the composition of poverty

reduction strategies in the ‘south’ so that economic growth, human capital development

and social protection are increasingly seen as the three elements of national

4

development strategies - a three pronged approach that increases national levels of

welfare, raises economic productivity and strengthens social cohesion.

This paper reviews the rise of social protection in developing countries. We argue that

the progress of social protection can be viewed as a quiet revolution.iv A decade ago

the idea that governments and international agencies would support development

policies that provided regular and reliable transfers to those in poverty would have been

seen as most improbable. Poor countries could not afford to waste scarce resources in

such a way, one would be creating a nation of welfare dependents and, anyway, even if

the idea had merit handouts would be stolen by corrupt officials and politicians.

Incrementally, such assertions have been confronted and partially resolved. With

minimal fanfare social protection has moved onto national and international policy

agendas. How times have changed.

There is a fast growing literature on social protection in developing countries, the vast

majority of which focused on the design and impact particular programmes or types of

programmes. It is a very valuable literature, as it is important to know what works in

poverty reduction. This paper focuses instead on embedding the rise of social protection

within global and regional trends. Our aim is to approach social protection in developing

countries not as a menu for a ‘social planner’, but in the round, and as emerging from,

and responding to, social and economic transformation. To draw on a well-worn analogy,

the paper aims to focus on the forest rather than the trees. Taking this vantage point will

help trace the contours of social protection, its diversity, and the factors that constrain its

expansion.

The conclusion argues for the energetic continuation of this quiet revolution both to

improve the welfare and prospects of the world’s poor and poorest and to strengthen

national and international solidarity and security. It identifies two knowledge frontiers for

researchers and policy makers. The first is finding mechanisms to scale up social

protection coverage in low income countries without turning it into an aid donor

development fad which is later cast aside. The second, about which very little is known,

is identifying approaches that will extend social protection into fragile states and regions

and difficult environments.

5

The paper is organised as follows. Section 1 explores the conceptual underpinnings of

social protection and competing approaches to development. Section 2 discusses key

drivers in the rise of social protection in developing countries. Section 3 examines

regional pathways in the extension of social protection. Section 4 discusses three

important factors determining the future course of social protection in developing

countries: the role of external actors, finance and delivery capacity, and politics. A final

section summarises the main conclusions.

2. Social protection and development: conceptual underpinnings

Social protection has been primarily understood as a policy framework describing “public

actions taken in response to levels of vulnerability, risk, and deprivation which are

deemed socially unacceptable within a given polity or society” (Conway, de Haan and

Norton, 2000). In advanced industrial countries, social protection is constituted by a set

of integrated institutions and programmes including social insurance, social assistance,

and employment protection and promotion. Since the early 1990s, and against a

background of economic crises, structural adjustment, and globalisation, social

protection has increasingly defined a distinct policy agenda in developing countries.

There are several distinguishing features of the emerging paradigm. In developing

countries, social protection has a strong focus on poverty reduction and on the poor and

poorest. It relies to an increasing extent on income transfers combined with access to

basic services and/or productive employment and asset building. It has a strong

‘productivist’ bent, in as much as it is expected to make a contribution to social and

economic development. Compared to social protection in advanced economies, social

protection in developing countries exhibits a plurality of providers, involving government

agencies, multilateral and bilateral international organisations, and national and

international NGOs.

In developing countries, social protection is grounded on a widely shared understanding

that poverty is multidimensional and persistent in time and across generations.v Within

this shared perspective, a primary cause of poverty is to be found in the constraints

faced by the poor in taking advantage of economic opportunity.vi Competing

explanations of the nature of these constraints have produced a diversity of approaches

to social protection in international development. Some define the main role of social

protection as lifting the constraints to human and economic development posed by

6

social risk, others locate it in ensuring the satisfaction of basic needs, and others

ground it in implementing a rights-based approach to human development. Competing

perspectives on risks or needs or rights support alternative views of what social

protection seeks to achieve (Munro, 2008).

This point can be illustrated by considering the different frameworks for social protection

that are used by leading multilateral organizations. The World Bank conceptualizes

social protection as social risk management and proposes policies that seek ‘to assist

individuals, households and communities in better managing income risks’ (Holzmann

and Jorgensen, 1999: 4). It moves beyond what it sees as ‘traditional’ social protection

by adding the goals of macroeconomic stability and financial market development. The

emphasis on risk assumes that vulnerability to hazards is a significant constraint on

economic and human development, and that efforts to reduce the likelihood of hazards,

or to ameliorate their effects on living standards, are essential for economic growth and

development.

The ILO understands social protection as arising from human rights. It is defined by

‘entitlement to benefits that society provides to individuals and households – through

public and collective measures – to protect against low or declining living standards

arising out of a number of basic risks and needs’ (van Ginneken 2006, p.11). The

international community acknowledged that social protection is a basic human right in

the Universal Declaration of Human Rights in 1948: ‘everyone has the right to a standard

of living adequate for the health and well-being of himself and of his family…’. The ILO’s

recent reformulation of its mission statement as involving efforts to ‘secure decent work

for women and men everywhere’ is an affirmation of its rights-based perspective

reflecting the Declaration’s commitment to extend social protection to all (ILO, 2001a:

39). A rights-based approach adds an additional element to social protection. It moves

social protection from a policy option to an obligation for states and international

governance structures.

The UN defines social protection as ‘a set of public and private policies and programmes

undertaken by societies in response to various contingencies to offset the absence or

substantial reduction of income from work; to provide assistance to families with children

as well as provide people with basic health care and housing’ (United Nations, 2000: 4).

It is underpinned by shared ‘fundamental values concerning acceptable levels and

7

security of access to income, livelihood, employment, health and education services,

nutrition and shelter” (ibid). In essence, this approach envisages social protection as

ensuring the satisfaction of basic human needs, as a precondition for human and

economic development.

The discussion above shows how social protection links up to competing development

perspectives. These conceptual underpinnings have implications for what practical

actions are (or are not) emphasized by different agencies. While many differences

remain between the definitions and meanings that different agencies adopt for social

protection, we believe that over the last ten years the overarching debate has moved on

from contrasting social risk management and basic human needs perspectives to a more

ambitious focus on capabilities.vii

3. The rise of social protection in development policy

Current interest in social protection among policy makers developed in the aftermath of

the structural adjustment policies of the 1980s and 1990s, and especially their failure to

promote growth and reduce poverty. This led to a realisation that a globalised economy

could produce dramatic downturns in human well-being, and to a better understanding of

the human and developmental costs associated with not having adequate social

protection policies and programmes in developing countries. More recently, the

Millennium Development Goals (MDGs) have focused attention on poverty and

vulnerability reduction.

Several factors explain the rise of social protection on the policy agenda, but the effects

of globalization and rapid economic transformation are the most important as they raise

the demand for social protection (Rodrik, 1997; 2001). The greater openness of

developing economies implies increased vulnerability to changes in global markets, and

a greater concentration of social and economic hazards on the less powerful

participants. In the 1980s and 1990s economic transformation unfolded at a rapid pace

in Latin America and East Asia. The 1980s were characterized by acute and sustained

economic and financial crises as well as structural adjustment in the economies of Latin

America. The financial crisis in 1997 affected in similar ways the countries in East Asia.

The transition economies underwent deep structural reforms and transformation. In all

cases, the outcomes of these changes were a rapid rise in poverty and vulnerability,

8

which laid bare glaring gaps in social protection. In the countries affected, the adverse

impacts of transformation were concentrated on the more vulnerable sectors. Rising

poverty and vulnerability, and the threat of conflict and social unrest they presage,

focused attention on strengthening social protection policies and programmes. Social

protection programmes were initiated in Brazil (Britto, 2008) and Indonesia

(Sumarto et al., 2008) in the wake of economic and financial crises. Increasing poverty

and vulnerability arising from globalization and economic transformation are therefore

key drivers for social protection.

Analytical work has facilitated an improved understanding of the costs associated with

the absence of effective social protection in developing countries. An extensive literature

is available measuring these costs in a variety of settings (Morduch, 1998; Dercon,

2005). There are large direct costs associated with economy, and sometimes, region-

wide, natural, economic, and political hazards. Gaps in social protection are responsible

for excess transient poverty and can also be responsible for chronic poverty, especially

in situations where the coping strategies available to those below or near the poverty line

are limited and, as a result, they are forced to adopt alternatives with detrimental long

run effects. Taking children out of school, cutting down on health care, sub-standard

nutrition, or less productive employment or crops, can push households into persistent

poverty. There are large long-term economic and human development losses associated

with not having adequate social protection, and consequently large gains to be captured

by establishing strong social protection institutions.viii

The International Development Goals, Millennium Declaration in 2000 and subsequent

agreement on the Millennium Development Goals has focused the attention of

international organisations, poor and rich country governments and the citizens and

celebrities of Europe and North America on poverty and vulnerability reduction more

than any other global initiative in the past (Hulme, 2007). Leaving aside an assessment

of the desirability of the enterprise,ix or the extent to which it sits together with ongoing

initiatives such as PRSPs and PRSs, the focus on poverty has encouraged the

extension of social protection in many developing countriesx.

4. Regional and national social protection trajectories

9

It is important to take account of the diversity in social protection evolution in developing

countries and appreciate that policies evolve out of specific national contexts. Historical

factors play a large part in explaining why policies differ from country to country. This

section draws out key features in regional pathways.

Latin America: Historically, social protection in Latin America focussed on workers in

formal employment. As a consequence, the majority of the region’s population was

excluded from public social protection until very recently. Acute economic crisis in the

early 1980s, followed by structural adjustment and economic liberalisation resulted in

rising vulnerability, poverty, and inequality. The initial response to this, in terms of social

protection, was to reform social insurance institutions for the formally employed (to

control fiscal deficits) and to mount fragmented, often externally financed, safety nets

and social fundsxi.

By the mid 1990s, it was clear that more comprehensive and permanent public

responses were needed. The move away from military and emergency governments

created democratic governments that had to engage with the strong popular demand for

social protection. This opened the way for a set of highly innovative, domestically

designed poverty and vulnerability reduction programmes. These programmes - Bolsa

Escola/Familia (Brazil), Progresa/Oportunidadesa (Mexico), and Chile Solidario (Chile) -

have mobilised regional and global interest in social protection policies.

The origins of Bolsa Escola are to be found in the innovative approach to

multidimensional and persistent poverty adopted by the Municipality of Campinas in

Brazil, in the mid 1990s, later extended to the rest of the country. Similarly, the

introduction of Mexico’s Progresaxii reflected both systematic learning from the politicised

and ineffective anti-poverty programmes which preceded its introduction in 1997, and

the need to address the human development deficits that underpin intergenerational

poverty in rural communities. Chile Solidario represents a new generation of integrated

anti-poverty programmes inspired by a capability approach. These new human

development programmes aimed to meet the short-term needs of the poorest

households, especially improved consumption and nutrition, and longer-term goals, such

as improved education and health. The Latin American dynamic for social protection is

now strongly national and regional.

10

South Asia: The deep-seated informal social protection of pre-colonial ‘Indian’ societies

was overlaid by colonial programmes of famine relief and public assistance for the

destitute. Subsequently, the social welfare initiatives of newly independent national

governments and later the programmes of national and international NGOs were added.

In many cases this means that social protection is now a mosaic of overlapping

programmes, often poorly funded and weakly implemented,. These are overseen by

welfare ministries that have limited capacity for policy analysis or evaluation compared to

ministries of finance.

While historically the political elites of South Asia have been ideologically inclined to

pursue social welfare through public policy there are differences between countries. Sri

Lanka has been more successful in financing and delivering social protection policies

than have Bangladesh and Pakistan. The southern states of India – Kerala, Karnataka,

Tamil Nadu and Andhra Pradesh – have been much more effective at implementing

national social protection policies (ranging from mid-day meals for schoolchildren to old

age pensions) than the north and north-eastern states. In Nepal much of the social

protection effort is a patchwork of aid donor and NGO projects.

In the 1990s and early 2000s there has been much donor-financed activity, ranging from

World Bank social funds, such as Sri Lanka’s poorly performing Janasaviya Trust Fund,

to BRAC’s Targeting the Ultra Poor Programme in Bangladesh (Hulme and Moore,

2008). Alongside these are government initiatives, for example old age pensions in

Bangladesh, India and Nepal and the Samurthi Programme in Sri Lanka.

Since 2004 India has taken a regional leadership role through its National Rural

Employment Guarantee Scheme (NREGS) and, in 2007, with the tabling of the

Unorganised Sector Worker’s Social Security Bill (USWSS). The NREGS is a social

assistance programme, seeking to ensure basic income security for vulnerable

households with economic capacity in rural areas. It extends, on a national scale, the

approach to social protection tested in the Maharashtra Employment Guarantee

Scheme. The USWSS Bill aims to incorporate informal urban workers (ie the majority)

into a basic social insurance scheme (Kannan, 2006). If successful these two schemes

will substantially reduce the insecurity of India’s vast rural and urban informal labour

forces. The early reports on the NREGS suggest that in relatively well governed states

11

the policy is being effectively implemented (Jacob and Varghese, 2006), while in poorly

governed states it has stalled (Louis, 2006).

The role and influence of international agencies and aid donors varies greatly in South

Asia and has reduced over the last ten years. Donors have substantial influence in

smaller more aid dependent countries (Nepal), less traction in larger countries

experiencing economic growth (Bangladesh) and minimal influence in India.

South-East and East Asia: In these regions there has been a common historical

reliance on family-based social protection, but with different policy pathways reflecting

different responses to rapid social transformation. Among the more economically

advanced countries (Korea, Taiwan, Malaysia, Thailand and Singapore) social insurance

is the core of social protection institutions. By contrast, among lower income countries in

South-East Asia, the Philippines and Indonesia in particular, the 1997 crisis undermined

social insurance. The immediate response to the financial crisis was the rapid expansion

of temporary safety nets which have become a more permanent feature in Indonesia

(Sumarto et al., 2008) shifting the country’s approach from social insurance to social

assistance.

Transition countries had a different starting point and evolution, especially China and

Vietnam, and their recent changes in social protection are primarily directed to

addressing problems associated with rapid economic transformation. In urban China,

economic liberalisation has led to a rapid decline in the strength and coverage of social

insurance based around productive units, and an equally rapid rise in social assistance

through the Minimum Living Standards Scheme (MLSS) covering more than 22 million

households by 2006. In rural China, the main social protection innovation has been the

introduction of mixed provision health insurance schemes but there are rising concerns

about the increasing vulnerability of rural dwellers (and the political threats that might

arise from this situation).

As a consequence of the diversity in initial conditions and institutions, the region has not

developed a dominant social protection model, but in general terms the emphasis in

higher income countries has been on strengthening social insurance institutions, while

the emphasis in lower income countries has been on social assistance. The latter also

appears to be the main focus of social protection in very low income countries, such as

12

Laos and Cambodia, where social protection is incipient, and restricted to fragmented

externally funded programmes (Cook and Kwon, 2006).

Sub-Saharan Africa: This region has a legacy of social protection institutions somewhat

akin to South Asia. Deeply-embedded, informal systems of social protection overlain by

a patchwork of colonial schemes and aid-financed social assistance programmes

(focused on humanitarian support), NGO initiatives and under funded, fragmented, and

partially implemented social insurance institutions for civil servants. The historical

evolution of social protection in much of the region is depressing. Emergency food aid,

famine relief, and humanitarian assistance have been central to social protection for

many countries since the 1970s.

Recently, a concern to shift from an emergency aid focus into more permanent social

protection programmes has led to the spread of aid-financed pilot cash transfers

schemes, targeted on the poorest and most vulnerable, and usually including human

development components. Such initiatives are underway in Zambia, Kenya, Malawi,

Uganda, Ghana, and Nigeria. The Protective Safety Nets Programme in Ethiopia

provides an example of a food security programme incorporating cash-based public

assistance components (Kebede, 2006).

The wealthier countries of the South, South Africa, Namibia, and Botswana, are the

exception, with a stronger social assistance focus relying on grants for vulnerable

groups, especially the elderly and children. These programmes have arisen out of

domestic political and technical debates. Social pensions in South Africa and Namibia

reflect the successful adaptation of colonial forms of social protection. Very recently,

social pensions have been introduced in Swaziland and Lesotho, perhaps signalling the

emergence of a distinct sub-regional approach to social protection. While the evolution

of social protection in South Africa is closely related to its political history, the country’s

experience shows the way in which a deeply embedded programme, the social pension,

has been adapted over time to address the changing nature of vulnerability, including

the rise in the incidence of HIV/AIDs and increased migration.

More than any other region in the world sub-Saharan Africa’s social protection trajectory

is likely to be heavily dependent on donor design and financing. Moreover, the capacity

of African intelligentsias (think-tanks, universities, policy advisors) to engage with

13

multilateral agencies and donors to debate policy choices and implementation options, is

lower than in other regionsxiii.

To sum up this brief review of regional trajectories, with few exceptions, developing

countries are addressing the vulnerability of the poor and poorest through strengthening

social assistance institutions and programmes. This applies to low income countries, in

which social insurance institutions are almost absent, but also in middle income

countries in Latin America and Southern Africa. This review suggests responses to

poverty and vulnerability will follow a range of pathways in different regions, depending

on the nature of their existing institutions (determining path dependence), level of

economic development (determining their fiscal space), and features of their economic

transformation (especially the interactions between longer term transformations such as

ageing and short term fractures such as transition or change in the development model).

5. The future of social protection in the developing world: actors, bottlenecks, and politics

(a) The role of external actors in the rise of social protection

The engagement of a wide range of providers and stakeholders in the extension of

social protection is a feature in developing countries. This section examines the role of

international actors.

Among multilaterals, the ILO has taken the lead in advocating and supporting social

protection in developing countries (Usui, 1994). Its tripartite governance system,

involving trade unions, employers associations and governments has proved effective in

gathering support for the extension of social protection for organised workers. However,

the growth of informality, and the relative decline of organised formal employment in

recent times, has been particularly challenging to these structures. Concerns with the

capacity of traditional approaches and institutions to extend social protection coverage to

informal workers, a majority in the South, has encouraged important shifts in

perspective. It has led to a new focus on ‘decent work’ as a framework for extending

basic rights to all workers (ILO, 2001b). There are direct linkages to social protection and

poverty reduction, in so far as decent work includes protection and provides a direct

escape from poverty. More recently, the ILO has advocated a basic package of social

14

protection measures among low income countries (ILO, 2006). This is an advocacy

exercise as, like most UN agencies, the ILO does not control funds that could be directly

allocated to this goal.

The World Bank developed a social protection strategy in the mid 1990s as a response

to the impact of structural adjustment on developing countries and the failure of its

‘social dimensions’ initiatives. The Bank’s Social Protection Group, initially focused on

labour market and pension reform, and safety nets, but more recently it has supported a

wider range of instruments including cash transfers.xiv The Bank is now a major player in

social protection, leveraging change through technical assistance and financial support.xv

Its role as a Bank restricts its social protection work in countries with high debt levels.

Partnerships with bilaterals, such as the Social Protection Trust Fund established by

DFID to support joint initiatives, provide a facility with which to influence policy

developments in these countries.

While the IMF has formally signed up to the goal of reducing poverty its interest in social

protection remains indirect, but highly significant. Its primary focus is short term macro-

economic and financial stability, nationally and internationally. In effect, it plays a key

role in ensuring that expenditure on social protection, and donor support, do not damage

the macro-economic environment of borrower countries. Depending on one’s analytical

position it can be seen as a villain (Deacon, 2007) or an incompetent (Stiglitz, 2002) in

terms of the impacts of its policies on poor people. At best, for proponents of social

protection, it is viewed as an obstacle to the financing of effective social protection.

Other parts of the UN family have adopted social protection policies, including UNDP,

UNICEF, WHO and WFP (United Nations, 2000). Their interests and influence tend to

vary with their mandate – UNICEF on child welfare, WHO on health issues and WFP on

hunger. Bilaterals like DFID, GTZ, and USAID are increasingly developing and

supporting social protection policies. DFID is becoming a major player through the

funding of social protection initiatives of multilaterals (DFID, 2005) and efforts at

leveraging policy change in the international system.

Except for humanitarian relief and assistance,xvi the adoption of social protection among

international NGOs has been slower. Receptiveness to the social protection agenda has

been greater among international NGOs committed to poverty reduction and advocating

15

policies directed at groups whose vulnerability arises from life course conditions such as

Help Age International, and Save the Children (Beales and German, 2006). NGOs

involved in the delivery of development programmes, on the other hand, have been

slower to adopt social protection. Interestingly, it is among the NGOs involved in

delivering emergency and humanitarian assistance that a receptiveness to social

protection, as a longer term response to conflict and emergency, is strongest (Harvey,

2005).

The focus on external actors in this section should not detract from acknowledging the

central role of national governments in formulating policies and coordinating the

extension of social protection. National governments also need to ensure that social

protection is integrated into national development strategies. Extending social protection

in developing countries is ultimately about establishing institutions that reflect and

strengthen solidarity and cooperation within a society. Externally imposed policy

transfers are unlikely to achieve these objectives.

(b) Bottlenecks: Financing and delivery capacity

Finance is rightly perceived as one of the main constraints on the expansion of social

protection, especially in low income countries. It is useful to distinguish between two

separate issues: (i) determining the level of financing required to ensure a minimum level

of social protection; and (ii) finding out how developing countries could finance this.

As regards the first issue, the ILO has performed simulations for low income countries in

Africa and Asia to indicate the resources required to provide a basic social assistance

package (Berendt, 2008). Their main findings are that the average cost of a basic

package, including a universal pension covering old age and disability and a child

benefit, would absorb around two to three percent of GDP.xvii Naturally, the cost of the

same package in different countries would differ in line with demographic,

macroeconomic, and fiscal conditions. At the same time, varying the level of the benefit

and targeting only the poor could significantly reduce the resources required. We could

also take a positive approach to determining the resources required by considering the

cost of existing social assistance programmes. With the exception of the Minimum Living

Standards Scheme in China, social assistance programmes in developing countries aim

to cover only a fraction of household consumption.xviii The targeted conditional cash

16

transfer programmes introduced in Latin America and the Caribbean absorb less than

one percent of GDP. So, around one to two percent of GDP appears to be a rough

estimate of the level of resources required to finance a basic social assistance package

for the poor in developing countries.

Financing this basic level of social assistance appears affordable for most developing

countries, but it is bound to be more difficult to achieve for low income countries with low

revenue mobilisation capacity. Uganda, for example, only manages to collect taxes

equivalent to 13 percent of GDP, and therefore allocating one percent of GDP for social

protection would require a substantial recasting of the budget. Economic growth could

generate additional resources, but social protection is needed most in countries with

poor records on growth. Aside from growth, there are three main options to be

considered. Firstly, raising tax revenues as a proportion of GDP through improvements

in the efficiency of tax collection agencies.xix Secondly, switching expenditure from poorly

performing poverty budget allocations. And thirdly, raising levels of official development

assistance to start up and sustain social protection programmes. If donors increase their

aid commitments, as many agreed at Monterrey in 2002 and Gleneagles in 2005, then

the resources would be available to provide long term financing for some low income

countries, such as Malawi and Mali, with limited growth prospects.

For most low income countries some combination of these measures must be pursued.

While foreign aid may be essential to meet the start up costs of programmes in the

medium and longer run, sustainable and effective social protection has to be financed

from domestic resources – donors cannot be expected, or relied upon, to finance this

key function in the medium and longer run. In middle income countries, there is greater

scope for expenditure switching, especially in countries devoting considerable resources

to poorly performing poverty reduction interventions or to subsidise financially

unsustainable and highly unequal social insurance schemes.xx

Any discussion of whether low income countries, can afford social protection must be

balanced by an examination of the costs of not providing effective social protection.

These generate long term restrictions on the development of human capital and

supportive institutions that become themselves a constraint on growth and development.

Secondly, in virtually all countries establishing social protection involves shifting the

financing mix from one based mainly on households and informal provision to a more

17

diversified mix. This is clearest in the case of health insurance institutions. In their

absence, out-of-pocket household expenditure on health care is often inefficient as well

as insufficient because responses to health shocks can crowd out investment in

preventative care and because they are rationed by available resources. Health

insurance instruments can improve the efficiency of households’ health expenditures

and make resources go further.

Delivery capacity limitations are a major challenge to the extension of social protection in

most low income countries. These apply at several points in the policy cycle, beginning

with the capacity to study, measure, and analyse poverty and vulnerability, the capacity

to design and implement appropriate policies, and the capacity to deliver and evaluate

social protection programmes. On the ground, a successful extension of social

protection will involve the horizontal integration of poverty researchers, policy analysts,

political scientists, financial experts, programme managers, information systems

analysts and developers, accountants, and field officers.

To date, developing these capacities in low income countries has rarely been an explicit

objective of policy makers, research institutes, or international organisations. In many

low income countries, government restrictions on recruitment and salaries, and

‘departmentalism’ make it unlikely that government agencies could create these

networks and ensure their integration in a reasonable time frame. However, we should

note that nothing succeeds like success – if pilot programmes work well and develop

momentum then it becomes easier to scale up implementation capacity even in

unfavourable environments.xxi This is an area in which technical assistance might be

prioritised, through donor support for organisational development and appropriate skills

training. There is also the potentially significant role of inter-governmental transfers of

information, knowledge, and know how across the developed and developing world, and

within the latter.xxii An alternative approach is to engage international NGOs or

consultancy companies to fill capacity gaps, especially in service delivery, but this

provides only a short term palliative not a longer term solution.

At present, and in the future, institutional partnershipsxxiii to devise, advocate and deliver

social protection seem likely. These can involve national governments, national and

international NGOs, bilateral and multilateral aid agencies and research institutes. Each

can contribute its strengths - national coverage, links to poor people, finance, analytical

18

and monitoring capacity etc. Such partnerships are not without their problems, however,

for example when a donor tries to impose its model of social protection on a recipient

government.

(c) Politics

Extending social protection in developing countries also requires a propitious political

environment in which demand for social protection can translate into effective

government responses. It is useful to make a distinction between the political conditions

needed for the introduction of social protection initiatives, and those required for the

sustainability of programmes.

Considering the adoption of social protection programmes, there is surprisingly little

research for developing countries. Public choice models of policy processes are perhaps

not very helpful in this context. In developing countries, and especially low income ones,

the shortcomings of median voter models of policy adoption are apparent. Voters are ill

informed about the relative advantages of policy options, and the policy promises of

politicians have little or no credibility (Keefer and Khemani, 2003). Patronage,

clientelism, and corruption undermine the basis for democratically competitive politics.

The political system is, as a result, less effective in aggregating voter preferences, than

in protecting and nurturing patron-client relationships.xxiv

This underscores the importance of factors exogenous to the domestic political system,

such as major disasters or crises, or the intervention of donors and NGOs, in forcing

social protection onto the political agenda. It is not surprising that the initiation of social

protection programmes often reflects a desire on the part of policy makers to counteract

real or perceived opposition to government policy, and the threat of social unrest. Social

protection programmes can play a very significant role in facilitating social and economic

transformation, especially where the associated losses are large and up front. A good

example is the introduction of Bolivia’s Bono Solidario, a non contributory pension

scheme. The government used this pension programme as a means of ensuring political

support for the privatisation of utilities, by promising to use the proceeds from

privatisation to fund these pensions (Gray-Molina, 1999). After successfully completing

the privatisation process the government suspended the pension entitlement, but later

reinstated it under public pressure (Barrientos, 2006). Similarly, the rapid expansion of

19

China’s Minimum Living Standards Scheme and Argentina’s Jefes y Jefas, were

prompted by rapidly rising unemployment and the threat of unrest.

The political conditions required for the political sustainability of social protection

programmes are less demanding, but these can be problematic. Discussing the spread

of income transfer programmes in Latin America, Britto (2008) notes how quickly political

support can be gathered for programmes that are perceived to be effective in reaching

those in poverty who are perceived to be efficient in the use of resources. Social

protection programmes can quickly build coalitions of support that can ensure their

sustainability. This can also be a disadvantage as social protection shows strong path

dependence, with the implication that it will be difficult to reform programmes or replace

them with better alternatives. Creating a political constituency supporting social

protection priorities is essential to securing sustainable social protection

6. Conclusions

This paper has reviewed the rapid unfolding of a ‘quiet revolution’ in developing

countries - the rise of social protection. In the space of a decade or so, social protection

has become one of the three main elements of national development strategies, with

growth and human development. Its conceptual basis has been clarified and extended,

from a single focus on risk to a broader focus on basic needs and capabilities. This is

also reflected in practice, with a rapid scaling up of programmes and policies that

combine income transfers with basic services, employment guarantees or asset building.

The increase in coverage is astonishing, and promises to make a significant contribution

to global poverty and vulnerability reduction.

The key global drivers behind the rise of social protection include: the impact of crises

and adjustment on poverty and vulnerability together with the ineffectiveness of short

term discretionary safety nets; the growing awareness that a globalised world implies

large costs associated with not having social protection; and, the international focus on

poverty reduction that brought about the MDGs. This was complemented with a

discussion of stylised pathways in the extension of social protection at regional and sub-

regional levels. The discussion on pathways demonstrated the diversity of social

protection responses, depending on the nature of existing institutions (determining path

dependence), level of economic development (defining the fiscal space), and the

20

features of social and economic transformation (determining the interactions between

longer term trends and short term fractures).

The future course of social protection, especially in low income countries, will be largely

determined by three factors. Firstly, the firming up of broad partnerships in which

external actors support national social protection strategies that are led and managed by

national government agencies. Secondly, success in finding innovative ways to reduce

financing constraints in the medium and longer-term. This involves reinforcing revenue

mobilisation in developing countries. Thirdly, strengthening demand for social

protection.xxv

Finally, we highlight the two important knowledge frontiers on which researchers and

policy makers should be focusing. The first, finding ways of scaling up successful social

protection programmes to the national level in low income countries is already a priority

and that should continue. The second, creating knowledge about social protection

programmes in fragile states and difficult environments, through mounting experiments

and pilot programmes, has been somewhat neglected. Social protection may have the

ability to meet the desperate, immediate needs of people living in fragile states whilst at

the same time supporting peace-building efforts and the demobilisation of militias,

contributing to improved prospects for national and international security and future

growth. This is a great challenge – but, the quiet revolution of social protection over the

last ten years suggests that this is an idea and practice that has not yet reached its limits

in contributing to human development in developing countries and beyond.

i This is acknowledged by many multilateral and bilateral organizations, national governments,

and NGOs (IADB, 2000; United Nations, 2000; ADB, 2001; ILO, 2001a; World Bank, 2001; HAI,

2003; DFID, 2005). The G8 Summit Declaration in June 2007 described social protection as ‘an

investment in a country’s economic future and a cost-effective way of fighting poverty’.

ii Social assistance includes programmes and policies supporting those in poverty and financed in

the main from tax revenues. In development, these programmes are a subset of poverty

reduction programmes, and among Washington institutions they are referred to as safety nets.

21

iii Admittedly, these are countries with large populations but there is now widespread coverage in

many smaller African, Asian, Caribbean and Latin American countries. For a database of social

protection programmes in developing countries see Barrientos and Holmes (2006) at

www.chronicpoverty.org.

iv Many of the programmes that have led to support for social protection are based on income

transfers.

v See Addison, Hulme and Kanbur (2008) for a detailed examination of the concept of poverty

dynamics.

vi This arises in part from their vulnerability to the impact of economic, social, and natural hazards.

vii This is reflected also in the objectives and design of social protection programmes. Chile

Solidario, an integrated extreme poverty eradication programme introduced in Chile in 2004 is

explicitly based on a capability perspective.

viii Behind these statistics are the harrowing accounts of individual human suffering that tens of

millions of people have experienced because social protection has not been available - hunger,

stunting, social stigma, lives constrained by withdrawal from education, disability and easily

preventable deaths.

ix For critiques see Clemens, Kenny and Moss (2004) and Saith (2006).

x See Barrientos and Hulme (2008) for more details.

xi Bolivia’s social fund was the model for scores of social funds that the World Bank set up around

the world.

xii In the mid 2000s this was renamed Oportunidades.

xiii In most African countries universities have been allowed to deteriorate and significant

proportions of educated people, sometimes the majority, have migrated.

xiv For a discussion of the evolution of social protection in the Bank see World Bank (2001). For a

discussion of social policy and social protection in the Bank see Hall (2007).

22

xv A recent review of social safety nets (social assistance) work by the World Bank in the period

2002-2006 concluded that 9 percent of all Bank projects (lending and analytical) involved safety

nets; that safety nets absorbed 3 percent of total Bank lending; and that safety nets absorb one

half of the social protection portfolio. (Milazzo and Grosh, 2007)

xvi Many NGOs have decades of experience in responding to emergency and humanitarian crises

and some are global leaders in this field.

xvii If a basic health insurance is added, it would absorb on average an extra 2 percent of GDP.

xviii Many programmes aim to transfer an additional twenty percent of household consumption, for

the average beneficiary household.

xix Warlters and Auriol argue convincingly that improvements in the efficiency of tax collection is

likely to be more effective than expanding the tax base as a means of raising revenues in low

income countries (Auriol and Warlters, 2002; Warlters and Auriol, 2005).

xx Sumarto et al. (2008) discuss how the expansion of social protection in Indonesia was financed

by switching resources away from petrol subsidies. Britto (2008) makes reference to Brazil’s

partially successful efforts to switch government subsidies from generous pensions for civil

servants to programmes like Bolsa Escola targeting the poor.

xxi See Rondinelli (1993) for an analysis and examples of scaling up administrative capacity.

Judith Tendler’s (1997) Good Governance in the Tropics provides a detailed case study.

xxii See Hulme (2007) for the argument that India is now in a position to make the development of

capacities for poverty analysis a major component of the work of the newly established India

International Development Agency (IIDA). There are clearly similar potentials for Brazil to publicly

(or privately) engage in social protection capacity development across Africa – what a

comparative advantage it would have in Lusophone Africa.

xxiii See Robinson, Hewitt and Harriss (1999) for a discussion of these relationships.

xxiv See Hickey (2008) for a detailed discussion of these issues in Africa.

xxv See Fukuda-Parr (2006) for a discussion of this issue.

23

References

ADB (2001). Social Protection Strategy. Manila: Asian Development Bank.

Addison, T., Hulme, D. & Kanbur, R. (forthcoming, 2008). Poverty Dynamics. Oxford:

OUP.

Auriol, E. & Warlters, M. (2002). Taxation base in developing countries. Mimeo,

Toulouse: ARQADE.

Barrientos, A. (2006). The missing piece of pension reform in Latin America: Poverty

reduction. Social Policy and Administration, 40(4), 369-384.

Barrientos, A. & Holmes, R. (2006). Social Assistance in Developing Countries

Database. Brighton: Institute of Development Studies.

Barrientos, A. & Hulme, D. (2008). Social Protection for the Poor and the Poorest:

Concepts, Policies and Politics. London: Palgrave.

Beales, S. & German, T. (2006). Situation analysis of social protection and cash

transfers in Africa. Report, London: Development Initiatives and Help Age

International.

Berendt, C. (2008) Can low income countries in sub-Saharan Africa afford basic social

protection? In A. Barrientos and D. Hulme (Eds), Social Protection for the Poor

and Poorest: Concepts, Policies and Politics. London: Palgrave.

Britto, T. F. D. (2008). The emergence and popularity of conditional cash transfer

programmes in Latin America. In A. Barrientos and D. Hulme (Eds), Social

Protection for the Poor and Poorest: Concepts, Policies and Politics. London:

Palgrave.

Clemens, M. A., Kenny, C. J. & Moss, T. J. (2004). The trouble with the MDGs:

Confronting expectations of aid and development success. CGD Working Paper

No. 40, Washington D.C.: Centre for Global Development.

Conway, T., de Haan, A., & Norton, A. (Eds) (2000). Social Protection: New Directions of

Donor Agencies. London: Department for International Development.

Cook, S. & Kwon, H-J. (2006). Economic development and social protection in East

Asia. Mimeo, Brighton: IDS.

24

Deacon, B. (2007). Global Social Policy and Global Governance. London: Sage.

Dercon, S. (ed.) (2005). Insurance Against Poverty. Oxford: Oxford University Press.

DFID (2005). Social transfers and chronic poverty: emerging evidence and the challenge

ahead. DFID Practice paper, London: DFID.

Fukuda-Parr, S. (2006). Millennium Development Goal 8: Indicators for international

human rights obligations? Human Rights Quarterly, 28, 966-997.

Gray-Molina, G. (1999). La economía política de reformas institucionales en Bolivia.

Working Paper R-350, Washington: Inter-American Development Bank.

HAI (2003). Population Ageing and Development: New Strategies for Social Protection.

Report, London: Help Age International.

Hall, A. (2007). Social Policies in the World Bank: Paradigms and Challenges. Global

Social Policy, 7(2), 151-175.

Harvey, P. (2005). Cash and vouchers in emergencies. London: Humanitarian Policy

Group, ODI.

Hickey, S. (2008). Conceptualising the politics of social protection in Africa. In A.

Barrientos and D. Hulme (Eds.), Social Protection for the Poor and the Poorest.

London: Palgrave.

Holzmann, R. & Jorgensen, S. (1999). Social Protection as Social Risk Management:

Conceptual Underpinnings for the Social Protection Strategy Paper. Journal of

International Development, 11, 1005-1027.

Hulme, D. (2007). Inclusive globalization: India’s role in tackling global poverty, Exim

Bank of India Annual Commencement Day lecture 2007, Mumbai, Exim Bank.

Hulme, D. & Moore, K. (2008). Assisting the poorest in Bangladesh: Learning from

BRAC’s ‘Targeting the Ultra Poor’ Programme. In A. Barrientos and D. Hulme

(Eds.), Social Protection for the Poor and the Poorest. London: Palgrave.

IADB (2000). Social Protection for Equity and Growth. Washington DC: Inter-American

Development Bank.

ILO (2001a). Social Security: A New Consensus. Geneva: International Labour Office.

ILO (2001b). Social Security: Issues, challenges and prospects. Report VI, Geneva:

International Labour Office.

25

ILO (2006). Social Security for All: Investing in global social and economic development.

Discussion Paper 16. Geneva: Social Security Department.

Jacob, A. & Varghese, R. (2006). NREGA implementation 1: Reasonable beginning in

Palakkad, Kerala. Economic and Political Weekly, 2 (December), 4943-4945.

Kannan, K. P. (2006). Employment and social security for the working poor: Two major

initiatives in India. New Dehli: National Commission for Enterprises in the

Unorganised Sector, Government of India.

Kebede, E. (2006). Moving from emergency food aid to predictable cash transfers:

Recent experience in Ethiopia. Development Policy Review, 24(5), 579-599.

Keefer, P. & Khemani, S. (2003). The Political Economy of Public Expenditures. Mimeo,

Washington DC: The World Bank.

Levy, S. (2006). Progress Against Poverty: Sustaining Mexico’s Progresa-Oportunidades

Program. Washington DC: Brookings Institution.

Louis, P. (2006). NREGA implementation 2: Birth pangs in Bihar. Economic and Political

Weekly, 2 (December), 4946-4947.

Milazzo, A. & Grosh, M. (2007). Social safety nets in World Bank lending and analytical

work: FY2002-2006. Social Protection Discussion Paper 0705, Washington DC:

The World Bank.

Morduch, J. (1998). Between the state and the market: Can informal insurance patch the

safety net? The World Bank Research Observer, 14(2), 187-207.

Munro, L. T. (2008). Risks, rights and needs: Compatible or contradictory bases for

social protection. In A. Barrientos and D. Hulme (Eds.), Social Protection for the

Poor and Poorest: Concepts, Policies and Politics. London: Palgrave.

Riddell, R (2007). Does Foreign Aid Really Work? Oxford: OUP.

Robinson, D, Hewitt, T & Harriss, J. (1999). Managing Development: Understanding

Inter-organisational relationships. London: Sage Publications.

Rodrik, D. (1997). Has globalization gone too far? Washington DC: Institute for

International Economics.

Rodrik, D. (2001). Por qué hay tanta inseguridad económica en America Latina? Revista

de la CEPAL, 73, 7-31.

26

Rondinelli, D (1993). Development Projects as Policy Experiments. London, Longman.

Saith, A. (2006). From universal values to Millennium Development Goals: Lost in

translation. Development and Change, 37(6), 1167-1199.

Stiglitz, J. (2002). Globalization and its Discontents. London: Allen Lane.

Sumarto, S., Suryahadi, A. & Bazzi, S. (2008). Indonesia’s social protection during and

after the crisis. In A. Barrientos & D. Hulme (eds.), Social Protection for the Poor

and Poorest: Concepts, Policies and Politics. London: Palgrave.

United Nations (2000). Enhancing social protection and reducing vulnerability in a

globalizing world. Report of the Secretary General to the Thirty-ninth Session

E/CN.5/2001/2. Washington DC: United Nations Economic and Social Council.

Usui, C. (1994). Welfare state development in a world system context: event history

analysis of first social insurance legislation among 60 countries, 1880-1960. In T.

Janoski and A. M. Hicks (Eds.), The Comparative Political Economy of the

Welfare State (pp.254-277). Cambridge: Cambridge University Press.

Van Ginneken, W. (2003). Extending social security: Policies for developing countries.

ESS Paper 13, Geneva: ILO.

Warlters, M. & Auriol, E. (2005). The Marginal Cost of Public Funds in Africa. Policy

Research Working Paper WPS 3679, Washington DC: The World Bank.

World Bank (2001). Social Protection Sector Strategy: from Safety Net to Springboard.

Sector Strategy Paper, Washington DC: The World Bank.

The Brooks World Poverty Institute (BWPI) creates and shares knowledge to help end global poverty. BWPI is multidisciplinary, researching poverty in both the rich and poor worlds. Our aim is to better understand why people are poor, what keeps them trapped in poverty and how they can be helped - drawing upon the very best international practice in research and policy making. The Brooks World Poverty Institute is chaired by Nobel Laureate, Professor Joseph E. Stiglitz.

Executive Director Professor Tony Addison Research Director Professor Michael Woolcock Associate Director Professor David Hulme Contact: Brooks World Poverty Institute The University of Manchester Humanities Bridgeford Street Building Oxford Road Manchester M13 9PL United Kingdom Email: bwpi@manchester.ac.uk www.manchester.ac.uk/bwpi

www.manchester.ac.uk/bwpi